Home Explore Help
Register Sign In
apache
/
hadoop
mirror of https://github.com/apache/hadoop.git
2
0
Fork 1
Files Issues 0 Wiki
Branch: branch-3.1.2
Branches Tags
hadoop
/
hadoop-tools
/
hadoop-resourceestimator
Wangda Tan 733d03485b Preparing for 3.1.2 release 6 years ago
..
src f2df6b8983 HADOOP-15025. Ensure singleton for ResourceEstimatorService. (Rui Li via Subru). 7 years ago
README.md 625039ef20 HADOOP-14840. Tool to estimate resource requirements of an application pipeline based on prior executions. (Rui Li via Subru). 7 years ago
pom.xml 733d03485b Preparing for 3.1.2 release 6 years ago

README.md

Resource Estimator Service

Resource Estimator Service can parse the history logs of production jobs, extract their resource consumption skylines in the past runs and predict their resource requirements for the new run.

Current Status

  • Support Hadoop YARN ResourceManager logs.
  • In-memory store for parsed history resource skyline and estimation.
  • A Linear Programming based estimator.
  • Provides REST interface to parse logs, query history store and estimations.

Upcoming features

  • UI to query history and edit and save estimations.
  • Persisent store implementation for store (either DB based or distributed key-value like HBase).
  • Integrate directly with the Hadoop YARN Reservation System to make a recurring reservation based on the estimated resources.

Refer to the design document for more details.